By Anders Rhod Gregersen
Deciding where to put a wind turbine is a lot like planting a tree: where you plant it is critical to its long term health and effectiveness.
For wind turbines, pinpointing the optimal location enables energy producers to maximize power generation at reduced energy costs. It’s something we at Vestas Wind Systems of Denmark have been working on for several years.
To do it, we mine Big Data and leverage predictive analytics for deep insights. We analyze literally petabytes of information, ranging from weather reports and tidal phases, to geospatial and sensor data, as well as satellite images and weather modeling research. In all, 160 factors are analyzed that influence location, plant performance and service life.
But the insights gained help us to do a lot more than determine prime locations. The analysis also helps us identify new markets for wind energy and helps our clients meet aggressive renewable energy goals. For a company like ours, that is committed to addressing the world’s energy requirements, it is of high importance to not only seek high-performance but also to generate an energy-efficient computing environment that offers a reduced carbon footprint through lower server energy consumption.
Our solution is based on IBM BigInsights software and the IBM Firestorm supercomputer. We then combines open source Apache Hadoop software with unique technologies and capabilities from IBM that enable us to process very large data sets by breaking up the data into chunks and coordinating the processing across a distributed environment.
Thus, we know exactly how the wind is distributed across potential sites, and can compare this data with the turbine design specifications to make sure the turbine can operate at optimal efficiency at that location.
What is also remarkable is the fact that the supercomputers and modeling software aren’t only used in the planning phase. Once the wind farm is up and running, they can keep crunching data to decide when the best time to schedule maintenance is, and what important information can be incorporated into future models to refine them.
As we believe that electricity from wind technologies will further increase during the next years, Big Data analytics is helping us to speed up this timeline and enter new markets to capitalize on growing demand for wind energy.